R E S EA R C H A R T I C L E
Abnormal Lateral Geniculate Nucleus and Optic Chiasm in Human Albinism Larissa Mcketton,1,2 Krista R. Kelly,2,3 and Keith A. Schneider1,2* 1
Department of Biology, York University, Toronto, Ontario, Canada Centre for Vision Research, York University, Toronto, Ontario, Canada 3 Department of Psychology, York University, Toronto, Ontario, Canada 2
ABSTRACT Our objective was to measure how the misrouting of retinal ganglion cell (RGC) fibers affects the organization of the optic chiasm and lateral geniculate nuclei (LGN) in human albinism. We compared the chiasmal structures and the LGN in both pigmented controls and patients with albinism by using high-resolution structural magnetic resonance imaging (MRI). We studied 12 patients with oculocutaneous albinism and 12 agematched pigmented controls. Using a 3T MRI scanner, we acquired a T1-weighted three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) image of the whole brain, oriented so that the optic nerves, chiasm, and tracts were in the same plane. We acquired multiple proton density-weighted images centered on the thalamus and midbrain, and averaged them to increase the signal, enabling precise manual
tracing of the anatomical boundaries of the LGN. Albinism patients exhibited significantly smaller diameters of the optic nerves, chiasm and tracts, and optic chiasm and LGN volume compared with controls (P < 0.001 for all). The reductions in chiasmal diameters in the albinism compared with the control group can be attributed to the abnormal crossing of optic fibers and the reduction of RGCs in the central retina. The volume of the LGN devoted to the center of the visual field may be reduced in albinism due to fewer RGCs representing the area where the fovea would normally lie. Our data may be clinically useful in addressing how genetic deficits compromise proper structural and functional development in the brain. J. Comp. Neurol. 522:2680–2687, 2014. C 2014 Wiley Periodicals, Inc. V
INDEXING TERMS: albinism; lateral geniculate nucleus; optic chaism; magnetic resonance imaging
Albinism, a genetic condition of hypopigmentation, is caused by melanocyte or melanin depletion (Kugelman and Van Scott, 1961). Melanin plays a significant role during typical ocular development. If melanin is absent, the fovea fails to develop properly, and neural connections between the retina and brain are altered (Fulton et al., 1978). As a result, those with albinism display a variety of ophthalmic deficits including reduced visual acuity (Fonda et al., 1971) and nystagmus (Rosenberg and Jabbari, 1987; Summers, 1996). In the human visual pathway, approximately half of retinal ganglion cell (RGC) axons cross at the chiasm (i.e., nasal retinal fibers), whereas the other half project ipsilaterally (i.e., temporal retinal fibers) to both lateral geniculate nuclei (LGN). The LGN is the primary subcortical visual relay nucleus (Kastner et al., 2006) and is organized into six interleaved monocular layers: the dorsal four parvocellular (P) layers and the dorsal two magnocellular (M) layers. In the albinism visual pathway, a large proporC 2014 Wiley Periodicals, Inc. V
2680
tion of RGC axons from the temporal retina are misrouted and cross at the optic chiasm, as seen in mammals such as cats (Creel, 1971), rats (Lund, 1965), tigers (Guillery and Kaas, 1973), and humans (Carroll et al., 1980; Creel et al., 1974, 1978; Hoffmann et al., 2005; Morland et al., 2002; von dem Hagen et al., 2008). For example, in human albinism there is an increase in crossing over of 70–85% of retinal fibers at the optic chiasm, whereas 15–30% of fibers project ipsilaterally. This misrouting leads to an abnormal arrangement of fibers projecting to the LGN, with a
Grant sponsor: The Dana Foundation; Grant sponsor: the Natural Sciences and Engineering Research Council of Canada (NSERC). *CORRESPONDENCE TO: Keith Schneider, Sherman Health Science Research Centre, 4700 Keele Street, Toronto, ON M3J 1P3, Canada. E-mail: [email protected] Received December 20, 2013; Revised February 16, 2014; Accepted February 18, 2014. DOI 10.1002/cne.23565 Published online April 12, 2014 in Wiley Online (wileyonlinelibrary.com)
The Journal of Comparative Neurology | Research in Systems Neuroscience 522:2680–2687 (2014)
Library
Abnormal visual system morphology in albinism
TABLE 1. Albinism participant characteristics Optical correction
Patient A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12
Age (yr) 48 46 19 21 48 44 56 22 47 45 17 29
Sex
Albinism classification
M F M M M F M F F F F F
OCA OCA-1 OCA-1 OCA-1a OCA-1 OCA-1 OCA OCA OCA OCA OCA-1 OCA-2
Sphere
Cylinder
Visual acuity
OD
OS
0.6 0.7 0.8 1.0 1.0 0.8 0.9 0.6 0.9 1.0 0.9 0.5
17.75 15.50 111.5
18.00 15.00 110.7
24.00 17.00 12.25 12.50 11.00 110.0 15.50 212.50
OD
Axis OS
25.75 25.00 23.25 22.50 23.00 22.75 No corrective lenses 22.25 20.50 22.25 16.25 23.25 22.50 12.25 22.50 21.75 14.50 22.75 24.25 12.00 – – 110.0 – – 15.75 23.50 23.50 212.25 23.00 23.00
OD
OS
010 004 –
163 177 –
016 175 009 017 – – 005 010
150 180 163 166 – – 180 169
Visual acuity was measured with the ETDRS LogMAR eye chart. Also included are the prescriptions of those who wore corrective lenses. Abbreviations: OCA, oculocutaneous albinism; OCA1, tyrosinase-related oculocutaneous albinism with no functional tyrosinase (OCA1a); OCA2, with functional tyrosinase enzyme.
predominant representation of just one eye in the contralateral hemisphere (Guillery et al., 1975). In animal models of albinism, cats (Guillery et al., 1974), minks (Sanderson et al., 1974), and monkeys (Guillery et al., 1984) exhibit fusion of LGN layers. Relatively little research has been conducted on the development of the LGN in humans with albinism. One postmortem study compared LGN morphology of a single human with albinism with a visually intact control (Guillery et al., 1975). The albinism subject displayed fusions among the M and P layers, as would be expected when neighboring layers receive a larger proportion of the crossing nasal fibers. These layers would then be innervated by the same eye, and in a sense be monocular. The LGN of the albinism subject appeared to be smaller, and abnormal in shape and orientation, displaying an elevated lateral tip (Guillery, 1986; von dem Hagen et al., 2008). However, this has been the only study to examine the LGN in human albinism (Guillery et al., 1975). Moreover, only one participant with albinism was compared with one control, and this was conducted post mortem. It is unclear whether these proportions are general properties of the LGN in albinism or due to extensive individual variability of the LGN (Hickey and Guillery, 1979). In the present study, we investigated the consequences of fiber misrouting on the development of the LGN in living human participants with albinism. Our primary focus was to compare LGN volume between pigmented controls and participants with albinism using high-resolution structural magnetic resonance imaging (MRI). Prior studies have described the visibility of the LGN using proton density-weighted (PD) MRI scans (Bridge et al., 2008; Fujita et al., 2001; Gupta et al.,
2009) that were able to differentiate thalamic nuclei in controls (Devlin et al., 2006). By acquiring multiple PDweighted images, we were able to precisely determine the anatomical boundaries of the human LGN and obtain an accurate measure of its volume. Based on a smaller LGN in a postmortem albinism subject (Guillery et al., 1975), we predict smaller LGN volumes in the albinism group compared with controls. Due to previous inconsistencies, our secondary focus was to confirm previously found decreases in optic nerve, chiasm, and tract measurements in human albinism. To accomplish this, we used a 3T MRI scanner with 1-mm3 isotropic voxel size to obtain higher resolution images than those previously reported. This research was conducted to add to the body of knowledge about the development and plasticity of the LGN in the human thalamus.
MATERIALS AND METHODS Participants Twelve patients with oculocutaneous albinism (OCA) with a mean age 6 standard error of the mean (SEM) of 37 6 4 years (seven female) were compared with 12 age-matched controls, 32 6 3 years (six female). All participants were in good health with no history of neurological disorders. Visual acuity measurements were performed by using an ETDRS LogMar eye chart (Precision Vision, La Salle, IL) on all participants. Table 1 includes albinism patient history and optical correction. All controls had normal or corrected-to-normal visual acuity (20/20) or better. All participants gave written informed consent, and the experimental protocol was approved by the York University Human Participants Review Committee.
The Journal of Comparative Neurology | Research in Systems Neuroscience
2681
L. McKetton et al.
Figure 1. Axial image of a T1-weighted scan from a control participant displaying the (A) right and (B) left optic nerves, optic chiasm widths in the mediolateral (X) and anteroposterior (Y) dimensions, and the (C) right and (D) left optic tracts.
Optic chiasm measurements Participants were scanned by using a Siemens (Erlangen, Germany) Trio 3T MRI scanner and 32-channel head coil in the Neuroimaging Laboratory at the Sherman Health Sciences Research Centre at York University. To reduce head motion, cushions were placed around the participants’ heads. A T1-weighted three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) image of the entire head with a 1-mm3 isotropic voxels was acquired with the following parameters: TR 5 1.9 s, TE 5 2.52 ms, 1-mm slice thickness, 256 3 256 matrix, 9 flip angle, parallel imaging acceleration factor (generalized autocalibrating partially parallel acquisition [GRAPPA]) 5 2. For all T1-weighted images, the optic nerves, optic chiasm, and optic tracts were reoriented in the same plane, resulting in a reformatted image that was parallel to the optic chiasm. Three independent raters blind to group membership manually traced the optic chiasm region-of-interest (ROI) using FSLView software (v. 4.1.8, http://www.fmrib.ox.ac.uk/fsl). The volume of the optic chiasm was measured on the median of these three masks for each participant. Measurements of the right and left optic nerves, right and left optic tracts, and optic chiasm widths in the mediolateral and anteroposterior planes were also performed three times each by four raters who were blind to group membership using the OsiriX length measurement tool (Rosset et al., 2004) (Fig. 1).
volume acquisition took 89 s. The PDweighted scans were interpolated to twice the resolution in each dimension, registered, corrected for motion between acquisitions using the flirt program from the FSL software package, and averaged. Three independent raters blind to group membership manually outlined the anatomical boundaries of the LGN three times each. The resulting ROIs were merged, creating a median mask for each rater, and the median of these was used to measure LGN volume (Fig. 2). All statistics were computed with SPSS Version 20 for Mac (IBM, Armonk, NY).
Whole brain volume Whole brain volumes were calculated from each participant by using volumetric segmentation that excluded the cerebellum and brainstem using Freesurfer software (http://surfer.nmr.mgh.harvard.edu).
RESULTS Intra- and inter-rater reliability Intraclass correlation coefficient (ICC) reliability analyses were conducted to ensure consistent measurements within (intra) and between (inter) raters. For chiasm measurements, all intra-rater ICCs were above 0.74, and all inter-rater ICCs were above 0.91. For LGN volume, all intra-rater ICCs were above 0.93, and all inter-rater ICCs were above 0.89. ICCs above 0.70 are considered to reflect a reliable measure (Cohen, 2001); thus our ICCs indicate that measurements were consistent both within and between raters.
Whole brain volume The whole brain volume was (mean 6 SEM) 1,097 6 25 cm3 for the control group and 1,027 6 29 cm3 for the albinism group (Table 2). These volumes were in the same range as previously reported for healthy controls (Allen et al., 2002). No significant group difference in whole brain volume was found, t(22) 5 21.81, P 5 0.084.
Optic nerve and optic tract diameter LGN volume A series of 30–40 PD-weighted images were acquired coronally with 19–48 slices, 1 or 2 mm thick, 256 3 256 matrix, 192-mm field of view, 0.75 3 0.75 mm2 in-plane resolution, TR 5 3 s, TE 5 26 ms, flip angle 5 120 , and parallel imaging acceleration factor (GRAPPA) 5 2. Each
2682
The respective mean diameters for the right and left optic nerves were 4.81 6 0.18 mm and 4.78 6 0.15 mm for the control group, and 3.81 6 0.01 mm and 3.77 6 0.08 mm for the albinism group. The respective mean diameters for the right and left optic tracts were 4.28 6 0.14 mm and 4.23 6 0.12 mm for the control
The Journal of Comparative Neurology | Research in Systems Neuroscience
Abnormal visual system morphology in albinism
group, and 3.29 6 0.08 mm and 3.40 6 0.09 mm for the albinism group. A 2 3 2 3 2 mixed model analysis of covariance (ANCOVA) with group (albinism and control) as the between-group variable, hemisphere (left and right) and diameter (optic nerve and optic tract) as the withinsubject variables, and brain volume and age as covariates revealed no significant group by hemisphere interaction, F(1,20) 5 0.90, P 5 0.35, or group by diameter interaction, F(1,20) 5 0.25, P 5 0.62. However, there was a significant main effect of group, F(1, 20) 5 39.0, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.013) revealed that the albinism group had significantly decreased left and right optic nerve and optic tract diameters compared with the control group (P < 0.001 for all) (Fig. 3).
367 6 15 mm3 in the control group and 252 6 12 mm3 in the albinism group. A 2 3 3 mixed model ANCOVA, with group (albinism and control) as the between-group variable, chiasm measurements (mediolateral width, anteroposterior width, and chiasm volume) as the within-subject variables, and brain volume and age as covariates revealed a significant group by chiasm measurement interaction, F(2,40) 5 35.4, P < 0.001, and a significant main effect of group, F(1,20) 5 35.6, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.017) revealed that the albinism group had significantly decreased chiasm volume and chiasm width in the mediolateral plane (P < 0.001), but not in the anteroposterior plane (P 5 0.10) (Fig. 4).
LGN volume
The respective right and left mean LGN volumes were 165.2 6 9.6 mm3 and 157.9 6 9.8 mm3 for the The optic chiasm width in the mediolateral plane was control group, and 105.6 6 8.7 mm3 and 98.2 6 6.0 13.01 6 0.35 mm in the control group and 9.65 6 0.31 mm mm3 for the albinism group. in the albinism group. Chiasm width in the anteroposterior A 2 3 2 mixed model ANCOVA with group (albinism plane was 4.37 6 0.22 mm in the control group and 4.01 6 and control) as the between-group variable, hemisphere 0.12 mm in the albinism group. Chiasm volume was (left and right) as the within-subject variable, and brain volume and age as covariates revealed no significant group by hemisphere interaction, F(1,20) 5 0.11, P 5 0.75, and no significant main effect of hemisphere, F(1, 20) 5 0.14, P 5 0.71. However, there was a significant main effect of group, F(1, 20) 5 22.1, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.025) revealed significantly decreased left (P < Figure 2. A: An example of an averaged coronal PD-weighted image slab that was interpolated to twice 0.001) and right (P 5 the resolution and half the voxel size in a control participant. B: Zoomed-in view of the right and left 0.001) LGN volumes in albiLGN. C: Manually traced right and left LGN regions of interest. nism participants compared
Optic chiasm width and volume
TABLE 2. Descriptive statistics for the control and albinism groups showing the mean (SEM) optic nerve diameter, optic chiasm width in the X and Y planes, optic chiasm volume, optic tract diameter, LGN volume, and whole brain volume. Optic nerve diameter (mm) Group Control Albinism P value
Optic chiasm width (mm)
Right
Left
X Plane
Y Plane
Optic chiasm volume (mm3)
4.8 (0.18) 3.8 (0.01)
Abnormal Lateral Geniculate Nucleus and Optic Chiasm in Human Albinism Larissa Mcketton,1,2 Krista R. Kelly,2,3 and Keith A. Schneider1,2* 1
Department of Biology, York University, Toronto, Ontario, Canada Centre for Vision Research, York University, Toronto, Ontario, Canada 3 Department of Psychology, York University, Toronto, Ontario, Canada 2
ABSTRACT Our objective was to measure how the misrouting of retinal ganglion cell (RGC) fibers affects the organization of the optic chiasm and lateral geniculate nuclei (LGN) in human albinism. We compared the chiasmal structures and the LGN in both pigmented controls and patients with albinism by using high-resolution structural magnetic resonance imaging (MRI). We studied 12 patients with oculocutaneous albinism and 12 agematched pigmented controls. Using a 3T MRI scanner, we acquired a T1-weighted three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) image of the whole brain, oriented so that the optic nerves, chiasm, and tracts were in the same plane. We acquired multiple proton density-weighted images centered on the thalamus and midbrain, and averaged them to increase the signal, enabling precise manual
tracing of the anatomical boundaries of the LGN. Albinism patients exhibited significantly smaller diameters of the optic nerves, chiasm and tracts, and optic chiasm and LGN volume compared with controls (P < 0.001 for all). The reductions in chiasmal diameters in the albinism compared with the control group can be attributed to the abnormal crossing of optic fibers and the reduction of RGCs in the central retina. The volume of the LGN devoted to the center of the visual field may be reduced in albinism due to fewer RGCs representing the area where the fovea would normally lie. Our data may be clinically useful in addressing how genetic deficits compromise proper structural and functional development in the brain. J. Comp. Neurol. 522:2680–2687, 2014. C 2014 Wiley Periodicals, Inc. V
INDEXING TERMS: albinism; lateral geniculate nucleus; optic chaism; magnetic resonance imaging
Albinism, a genetic condition of hypopigmentation, is caused by melanocyte or melanin depletion (Kugelman and Van Scott, 1961). Melanin plays a significant role during typical ocular development. If melanin is absent, the fovea fails to develop properly, and neural connections between the retina and brain are altered (Fulton et al., 1978). As a result, those with albinism display a variety of ophthalmic deficits including reduced visual acuity (Fonda et al., 1971) and nystagmus (Rosenberg and Jabbari, 1987; Summers, 1996). In the human visual pathway, approximately half of retinal ganglion cell (RGC) axons cross at the chiasm (i.e., nasal retinal fibers), whereas the other half project ipsilaterally (i.e., temporal retinal fibers) to both lateral geniculate nuclei (LGN). The LGN is the primary subcortical visual relay nucleus (Kastner et al., 2006) and is organized into six interleaved monocular layers: the dorsal four parvocellular (P) layers and the dorsal two magnocellular (M) layers. In the albinism visual pathway, a large proporC 2014 Wiley Periodicals, Inc. V
2680
tion of RGC axons from the temporal retina are misrouted and cross at the optic chiasm, as seen in mammals such as cats (Creel, 1971), rats (Lund, 1965), tigers (Guillery and Kaas, 1973), and humans (Carroll et al., 1980; Creel et al., 1974, 1978; Hoffmann et al., 2005; Morland et al., 2002; von dem Hagen et al., 2008). For example, in human albinism there is an increase in crossing over of 70–85% of retinal fibers at the optic chiasm, whereas 15–30% of fibers project ipsilaterally. This misrouting leads to an abnormal arrangement of fibers projecting to the LGN, with a
Grant sponsor: The Dana Foundation; Grant sponsor: the Natural Sciences and Engineering Research Council of Canada (NSERC). *CORRESPONDENCE TO: Keith Schneider, Sherman Health Science Research Centre, 4700 Keele Street, Toronto, ON M3J 1P3, Canada. E-mail: [email protected] Received December 20, 2013; Revised February 16, 2014; Accepted February 18, 2014. DOI 10.1002/cne.23565 Published online April 12, 2014 in Wiley Online (wileyonlinelibrary.com)
The Journal of Comparative Neurology | Research in Systems Neuroscience 522:2680–2687 (2014)
Library
Abnormal visual system morphology in albinism
TABLE 1. Albinism participant characteristics Optical correction
Patient A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12
Age (yr) 48 46 19 21 48 44 56 22 47 45 17 29
Sex
Albinism classification
M F M M M F M F F F F F
OCA OCA-1 OCA-1 OCA-1a OCA-1 OCA-1 OCA OCA OCA OCA OCA-1 OCA-2
Sphere
Cylinder
Visual acuity
OD
OS
0.6 0.7 0.8 1.0 1.0 0.8 0.9 0.6 0.9 1.0 0.9 0.5
17.75 15.50 111.5
18.00 15.00 110.7
24.00 17.00 12.25 12.50 11.00 110.0 15.50 212.50
OD
Axis OS
25.75 25.00 23.25 22.50 23.00 22.75 No corrective lenses 22.25 20.50 22.25 16.25 23.25 22.50 12.25 22.50 21.75 14.50 22.75 24.25 12.00 – – 110.0 – – 15.75 23.50 23.50 212.25 23.00 23.00
OD
OS
010 004 –
163 177 –
016 175 009 017 – – 005 010
150 180 163 166 – – 180 169
Visual acuity was measured with the ETDRS LogMAR eye chart. Also included are the prescriptions of those who wore corrective lenses. Abbreviations: OCA, oculocutaneous albinism; OCA1, tyrosinase-related oculocutaneous albinism with no functional tyrosinase (OCA1a); OCA2, with functional tyrosinase enzyme.
predominant representation of just one eye in the contralateral hemisphere (Guillery et al., 1975). In animal models of albinism, cats (Guillery et al., 1974), minks (Sanderson et al., 1974), and monkeys (Guillery et al., 1984) exhibit fusion of LGN layers. Relatively little research has been conducted on the development of the LGN in humans with albinism. One postmortem study compared LGN morphology of a single human with albinism with a visually intact control (Guillery et al., 1975). The albinism subject displayed fusions among the M and P layers, as would be expected when neighboring layers receive a larger proportion of the crossing nasal fibers. These layers would then be innervated by the same eye, and in a sense be monocular. The LGN of the albinism subject appeared to be smaller, and abnormal in shape and orientation, displaying an elevated lateral tip (Guillery, 1986; von dem Hagen et al., 2008). However, this has been the only study to examine the LGN in human albinism (Guillery et al., 1975). Moreover, only one participant with albinism was compared with one control, and this was conducted post mortem. It is unclear whether these proportions are general properties of the LGN in albinism or due to extensive individual variability of the LGN (Hickey and Guillery, 1979). In the present study, we investigated the consequences of fiber misrouting on the development of the LGN in living human participants with albinism. Our primary focus was to compare LGN volume between pigmented controls and participants with albinism using high-resolution structural magnetic resonance imaging (MRI). Prior studies have described the visibility of the LGN using proton density-weighted (PD) MRI scans (Bridge et al., 2008; Fujita et al., 2001; Gupta et al.,
2009) that were able to differentiate thalamic nuclei in controls (Devlin et al., 2006). By acquiring multiple PDweighted images, we were able to precisely determine the anatomical boundaries of the human LGN and obtain an accurate measure of its volume. Based on a smaller LGN in a postmortem albinism subject (Guillery et al., 1975), we predict smaller LGN volumes in the albinism group compared with controls. Due to previous inconsistencies, our secondary focus was to confirm previously found decreases in optic nerve, chiasm, and tract measurements in human albinism. To accomplish this, we used a 3T MRI scanner with 1-mm3 isotropic voxel size to obtain higher resolution images than those previously reported. This research was conducted to add to the body of knowledge about the development and plasticity of the LGN in the human thalamus.
MATERIALS AND METHODS Participants Twelve patients with oculocutaneous albinism (OCA) with a mean age 6 standard error of the mean (SEM) of 37 6 4 years (seven female) were compared with 12 age-matched controls, 32 6 3 years (six female). All participants were in good health with no history of neurological disorders. Visual acuity measurements were performed by using an ETDRS LogMar eye chart (Precision Vision, La Salle, IL) on all participants. Table 1 includes albinism patient history and optical correction. All controls had normal or corrected-to-normal visual acuity (20/20) or better. All participants gave written informed consent, and the experimental protocol was approved by the York University Human Participants Review Committee.
The Journal of Comparative Neurology | Research in Systems Neuroscience
2681
L. McKetton et al.
Figure 1. Axial image of a T1-weighted scan from a control participant displaying the (A) right and (B) left optic nerves, optic chiasm widths in the mediolateral (X) and anteroposterior (Y) dimensions, and the (C) right and (D) left optic tracts.
Optic chiasm measurements Participants were scanned by using a Siemens (Erlangen, Germany) Trio 3T MRI scanner and 32-channel head coil in the Neuroimaging Laboratory at the Sherman Health Sciences Research Centre at York University. To reduce head motion, cushions were placed around the participants’ heads. A T1-weighted three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) image of the entire head with a 1-mm3 isotropic voxels was acquired with the following parameters: TR 5 1.9 s, TE 5 2.52 ms, 1-mm slice thickness, 256 3 256 matrix, 9 flip angle, parallel imaging acceleration factor (generalized autocalibrating partially parallel acquisition [GRAPPA]) 5 2. For all T1-weighted images, the optic nerves, optic chiasm, and optic tracts were reoriented in the same plane, resulting in a reformatted image that was parallel to the optic chiasm. Three independent raters blind to group membership manually traced the optic chiasm region-of-interest (ROI) using FSLView software (v. 4.1.8, http://www.fmrib.ox.ac.uk/fsl). The volume of the optic chiasm was measured on the median of these three masks for each participant. Measurements of the right and left optic nerves, right and left optic tracts, and optic chiasm widths in the mediolateral and anteroposterior planes were also performed three times each by four raters who were blind to group membership using the OsiriX length measurement tool (Rosset et al., 2004) (Fig. 1).
volume acquisition took 89 s. The PDweighted scans were interpolated to twice the resolution in each dimension, registered, corrected for motion between acquisitions using the flirt program from the FSL software package, and averaged. Three independent raters blind to group membership manually outlined the anatomical boundaries of the LGN three times each. The resulting ROIs were merged, creating a median mask for each rater, and the median of these was used to measure LGN volume (Fig. 2). All statistics were computed with SPSS Version 20 for Mac (IBM, Armonk, NY).
Whole brain volume Whole brain volumes were calculated from each participant by using volumetric segmentation that excluded the cerebellum and brainstem using Freesurfer software (http://surfer.nmr.mgh.harvard.edu).
RESULTS Intra- and inter-rater reliability Intraclass correlation coefficient (ICC) reliability analyses were conducted to ensure consistent measurements within (intra) and between (inter) raters. For chiasm measurements, all intra-rater ICCs were above 0.74, and all inter-rater ICCs were above 0.91. For LGN volume, all intra-rater ICCs were above 0.93, and all inter-rater ICCs were above 0.89. ICCs above 0.70 are considered to reflect a reliable measure (Cohen, 2001); thus our ICCs indicate that measurements were consistent both within and between raters.
Whole brain volume The whole brain volume was (mean 6 SEM) 1,097 6 25 cm3 for the control group and 1,027 6 29 cm3 for the albinism group (Table 2). These volumes were in the same range as previously reported for healthy controls (Allen et al., 2002). No significant group difference in whole brain volume was found, t(22) 5 21.81, P 5 0.084.
Optic nerve and optic tract diameter LGN volume A series of 30–40 PD-weighted images were acquired coronally with 19–48 slices, 1 or 2 mm thick, 256 3 256 matrix, 192-mm field of view, 0.75 3 0.75 mm2 in-plane resolution, TR 5 3 s, TE 5 26 ms, flip angle 5 120 , and parallel imaging acceleration factor (GRAPPA) 5 2. Each
2682
The respective mean diameters for the right and left optic nerves were 4.81 6 0.18 mm and 4.78 6 0.15 mm for the control group, and 3.81 6 0.01 mm and 3.77 6 0.08 mm for the albinism group. The respective mean diameters for the right and left optic tracts were 4.28 6 0.14 mm and 4.23 6 0.12 mm for the control
The Journal of Comparative Neurology | Research in Systems Neuroscience
Abnormal visual system morphology in albinism
group, and 3.29 6 0.08 mm and 3.40 6 0.09 mm for the albinism group. A 2 3 2 3 2 mixed model analysis of covariance (ANCOVA) with group (albinism and control) as the between-group variable, hemisphere (left and right) and diameter (optic nerve and optic tract) as the withinsubject variables, and brain volume and age as covariates revealed no significant group by hemisphere interaction, F(1,20) 5 0.90, P 5 0.35, or group by diameter interaction, F(1,20) 5 0.25, P 5 0.62. However, there was a significant main effect of group, F(1, 20) 5 39.0, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.013) revealed that the albinism group had significantly decreased left and right optic nerve and optic tract diameters compared with the control group (P < 0.001 for all) (Fig. 3).
367 6 15 mm3 in the control group and 252 6 12 mm3 in the albinism group. A 2 3 3 mixed model ANCOVA, with group (albinism and control) as the between-group variable, chiasm measurements (mediolateral width, anteroposterior width, and chiasm volume) as the within-subject variables, and brain volume and age as covariates revealed a significant group by chiasm measurement interaction, F(2,40) 5 35.4, P < 0.001, and a significant main effect of group, F(1,20) 5 35.6, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.017) revealed that the albinism group had significantly decreased chiasm volume and chiasm width in the mediolateral plane (P < 0.001), but not in the anteroposterior plane (P 5 0.10) (Fig. 4).
LGN volume
The respective right and left mean LGN volumes were 165.2 6 9.6 mm3 and 157.9 6 9.8 mm3 for the The optic chiasm width in the mediolateral plane was control group, and 105.6 6 8.7 mm3 and 98.2 6 6.0 13.01 6 0.35 mm in the control group and 9.65 6 0.31 mm mm3 for the albinism group. in the albinism group. Chiasm width in the anteroposterior A 2 3 2 mixed model ANCOVA with group (albinism plane was 4.37 6 0.22 mm in the control group and 4.01 6 and control) as the between-group variable, hemisphere 0.12 mm in the albinism group. Chiasm volume was (left and right) as the within-subject variable, and brain volume and age as covariates revealed no significant group by hemisphere interaction, F(1,20) 5 0.11, P 5 0.75, and no significant main effect of hemisphere, F(1, 20) 5 0.14, P 5 0.71. However, there was a significant main effect of group, F(1, 20) 5 22.1, P < 0.001. Post hoc pairwise comparisons (Bonferroni-adjusted a 5 0.025) revealed significantly decreased left (P < Figure 2. A: An example of an averaged coronal PD-weighted image slab that was interpolated to twice 0.001) and right (P 5 the resolution and half the voxel size in a control participant. B: Zoomed-in view of the right and left 0.001) LGN volumes in albiLGN. C: Manually traced right and left LGN regions of interest. nism participants compared
Optic chiasm width and volume
TABLE 2. Descriptive statistics for the control and albinism groups showing the mean (SEM) optic nerve diameter, optic chiasm width in the X and Y planes, optic chiasm volume, optic tract diameter, LGN volume, and whole brain volume. Optic nerve diameter (mm) Group Control Albinism P value
Optic chiasm width (mm)
Right
Left
X Plane
Y Plane
Optic chiasm volume (mm3)
4.8 (0.18) 3.8 (0.01)
